Machine learning(ML)has powerful nonlinear processing and multivariate learning capabilities,so it has been widely utilised in the fatigue field.However,most ML methods are inexplicable black-box models that are diffi...Machine learning(ML)has powerful nonlinear processing and multivariate learning capabilities,so it has been widely utilised in the fatigue field.However,most ML methods are inexplicable black-box models that are difficult to apply in engineering practice.Symbolic regression(SR)is an interpretable machine learning method for determining the optimal fitting equation for datasets.In this study,domain knowledge-guided SR was used to determine a new fatigue crack growth(FCG)rate model.Three terms of the variable subtree ofΔK,R-ratio,andΔK_(th)were obtained by analysing eight traditional semi-empirical FCG rate models.Based on the FCG rate test data from other literature,the SR model was constructed using Al-7055-T7511.It was subsequently extended to other alloys(Ti-10V-2Fe-3Al,Ti-6Al-4V,Cr-Mo-V,LC9cs,Al-6013-T651,and Al-2324-T3)using multiple linear regression.Compared with the three semi-empirical FCG rate models,the SR model yielded higher prediction accuracy.This result demonstrates the potential of domain knowledge-guided SR for building the FCG rate model.展开更多
The electromagnetic field of a human heart system is a bioelectromagnetic field. Electrocardiography (ECG) and magnetocardiography (MCG) are both carriers of electromagnetic information about the cardiac system, a...The electromagnetic field of a human heart system is a bioelectromagnetic field. Electrocardiography (ECG) and magnetocardiography (MCG) are both carriers of electromagnetic information about the cardiac system, and they are nonstationary signals. In this study, ECG and MCG data from healthy subjects are acquired; the MCG data are captured using a high-Tc radio frequency superconducting quantum interference device (HTc rf SQUIDs) and the QRS complexes in these data are analysed by the evolutionary spectrum analysis method. The results show that the quality factor Q and the central frequency fz of the QRS complex evolutionary spectrum are the characteristic parameters (CHPs) of ECG and MCG in the time-frequency domain. The confidence intervals of the mean values of the CHPs are estimated by the Student t distribution method in mathematical statistics. We believe that there are threshold ranges of the mean values of Q and fz for healthy subjects. We have postulated the following criterion: if the mean values of CHPs are in the proper ranges, the cardiac system is in a normal condition and it possesses the capability of homeostasis. In contrast, if the mean values of the CHPs do not lie in the proper ranges, the homeostasis of the cardiac system is lacking and some cardiac disease may follow. The results and procedure of MCG CHPs in the study afford a technological route for the application of HTc rf SQUIDs in cardiology.展开更多
Fatigue crack propagation rate,da/dN,and threshold stress intensity range,ΔK_(th),of steel 20CrMnMo containing low carbon martensite or low carbon martensite/bainite(LCM/B) duplex structure,obtained by oil quenching ...Fatigue crack propagation rate,da/dN,and threshold stress intensity range,ΔK_(th),of steel 20CrMnMo containing low carbon martensite or low carbon martensite/bainite(LCM/B) duplex structure,obtained by oil quenching and austempered at 360℃,have been measured using specimens under four-point bending.Observation was also made of the crack path and fracture morphology with relation to microstructure.The formation of LCM/B duplex structure,which caused by 20% lower bainite after short-time isothermal treatment,may re- markably increase ΔK_(th)and considerably decrease da/dN.The effect of microstructure on da/dN and ΔK_(th)was discussed with the emphasis on the crack propagation path.展开更多
基金Supported by Sichuan Provincial Science and Technology Program(Grant No.2022YFH0075)Opening Project of State Key Laboratory of Performance Monitoring and Protecting of Rail Transit Infrastructure(Grant No.HJGZ2021113)Independent Research Project of State Key Laboratory of Traction Power(Grant No.2022TPL_T03).
文摘Machine learning(ML)has powerful nonlinear processing and multivariate learning capabilities,so it has been widely utilised in the fatigue field.However,most ML methods are inexplicable black-box models that are difficult to apply in engineering practice.Symbolic regression(SR)is an interpretable machine learning method for determining the optimal fitting equation for datasets.In this study,domain knowledge-guided SR was used to determine a new fatigue crack growth(FCG)rate model.Three terms of the variable subtree ofΔK,R-ratio,andΔK_(th)were obtained by analysing eight traditional semi-empirical FCG rate models.Based on the FCG rate test data from other literature,the SR model was constructed using Al-7055-T7511.It was subsequently extended to other alloys(Ti-10V-2Fe-3Al,Ti-6Al-4V,Cr-Mo-V,LC9cs,Al-6013-T651,and Al-2324-T3)using multiple linear regression.Compared with the three semi-empirical FCG rate models,the SR model yielded higher prediction accuracy.This result demonstrates the potential of domain knowledge-guided SR for building the FCG rate model.
基金supported by the National Natural Science Foundation of China (Grant No. 607710003)
文摘The electromagnetic field of a human heart system is a bioelectromagnetic field. Electrocardiography (ECG) and magnetocardiography (MCG) are both carriers of electromagnetic information about the cardiac system, and they are nonstationary signals. In this study, ECG and MCG data from healthy subjects are acquired; the MCG data are captured using a high-Tc radio frequency superconducting quantum interference device (HTc rf SQUIDs) and the QRS complexes in these data are analysed by the evolutionary spectrum analysis method. The results show that the quality factor Q and the central frequency fz of the QRS complex evolutionary spectrum are the characteristic parameters (CHPs) of ECG and MCG in the time-frequency domain. The confidence intervals of the mean values of the CHPs are estimated by the Student t distribution method in mathematical statistics. We believe that there are threshold ranges of the mean values of Q and fz for healthy subjects. We have postulated the following criterion: if the mean values of CHPs are in the proper ranges, the cardiac system is in a normal condition and it possesses the capability of homeostasis. In contrast, if the mean values of the CHPs do not lie in the proper ranges, the homeostasis of the cardiac system is lacking and some cardiac disease may follow. The results and procedure of MCG CHPs in the study afford a technological route for the application of HTc rf SQUIDs in cardiology.
文摘Fatigue crack propagation rate,da/dN,and threshold stress intensity range,ΔK_(th),of steel 20CrMnMo containing low carbon martensite or low carbon martensite/bainite(LCM/B) duplex structure,obtained by oil quenching and austempered at 360℃,have been measured using specimens under four-point bending.Observation was also made of the crack path and fracture morphology with relation to microstructure.The formation of LCM/B duplex structure,which caused by 20% lower bainite after short-time isothermal treatment,may re- markably increase ΔK_(th)and considerably decrease da/dN.The effect of microstructure on da/dN and ΔK_(th)was discussed with the emphasis on the crack propagation path.